This PDF file contains the front matter associated with SPIE Proceedings Volume 12949, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

The talk will begin with a brief introduction of structural health monitoring (SHM) which has been attracting intensive attention since early 1990s. An essential difference between sensor centric SHM and nondestructive evaluation (NDE) will be highlighted. Advances in smart sensors powered by energy harvesting via ambient vibrations will be exemplified by two practical case studies. Recent advances in computer vision based SHM techniques using optical non-contact sensors with machine learning to detect impact loading and barely visible impact damage (BVID) in composite panels will be discussed in details. Finally, the digital twin framework under digital transformation and artificial intelligence (AI) is gaining potential to pave the way for future aircraft health monitoring.

Nonlinear ultrasonic guided waves have attracted increasing attention in non-destructive evaluation (NDE) due to their advantages over linear ultrasonic guided waves, e.g. potential to be a baseline-free technique for damage detection and higher sensitivity to incipient damage in structures. This study presents an experimental investigation of using nonlinear ultrasonic guided waves to evaluate damage when reinforced concrete beams are subjected to high temperatures. In this study, the reinforced concrete beam specimens are heated up to high temperatures in a furnace and it shows that the high temperatures induce debonding between rebar and concrete. Ultrasonic guided wave inspection is carried out at different high temperatures. When the ultrasonic guided wave interacts with the debonding between rebar and concrete, the second harmonic, which is one of the nonlinear features of ultrasonic guided waves, is generated due to contact acoustic nonlinearity (CAN). In this study, a nonlinear parameter is defined to quantify the second harmonic generation due to different levels of debonding damage in the reinforced concrete beams at different temperatures. The relationship between the debonding damage caused by the high temperature and the characteristics of the nonlinear parameter is studied. The results show that the nonlinear parameter can provide an indication of the severity of the debonding damage, and can be used to monitor the growth of the debonding damage in the reinforced concrete beams at high temperatures.

Coastal regions, particularly in the southeastern United States, are consistently confronted with the ongoing threat of hurricane-induced damage to their buildings and civil infrastructure. Consequently, there is a pressing need to concentrate efforts on the evaluation and prediction of structural integrity and reliability in such environments. This is paramount for minimizing losses and enhancing public safety in the face of these challenging climatic conditions. Current structural health monitoring systems are typically customized for specific buildings, rendering them excessively expensive and impractical for residential structures. This research presents a comprehensive feasibility study for an economical yet efficient system designed to anticipate potential failures and assess the safety and reliability of residential buildings. The proposed system employs integrated piezoelectric sensors to monitor alterations in the structural and material characteristics of building components. The collected sensor data may be transmitted to a mobile application using a WiFi or Bluetooth system. To validate the functionality of this innovative system, a proof-of-concept prototype building was constructed utilizing additive manufacturing, featuring integrated piezoelectric sensors. The system underwent experimental testing under base excitation at various frequencies, revealing distinct output variations at different locations. This substantiates the feasibility of employing integrated piezoelectric sensors within structural buildings for effective structural health monitoring. The collected data will serve as a foundational resource for accurately estimating the building’s reliability in anticipation of future hurricane events.

In the aerospace industry, the demand for precise non-destructive testing has grown to ensure the safety of aircraft and spacecraft. Among these, laser ultrasonic non-destructive testing is emerging as a favored technique in structural health monitoring due to its ability to inspect from a distance without requiring a medium.

Laser mirror scanners (LMS), which are composed of two galvanometers to direct a laser beam onto an inspection target, have become increasingly popular in laser scanning systems. Their advantages include being more cost-effective, faster in beam steering, and more compact in size compared to systems that utilize linear servo motors.

However, when considering mobile inspection systems for on-site inspection, there's a challenge in determining the initial steering point (or the origin point) of the LMS. This can be risky as there's the potential to direct the laser beam to an unintended location.

In this study, a calibration technique is investigated that utilizes a LiDAR camera to capture a three-dimensional point cloud of the specimen, enabling the LMS to accurately direct the laser beam to the desired location.

Masonry structures constitute a significant portion of the world's cultural heritage, and ensuring their structural and seismic safety presents formidable challenges. These challenges include highly nonlinear mechanical behavior, complex failure mechanisms, heterogeneous material properties, and intricate modeling requirements. Structural health monitoring (SHM) has emerged as a pivotal approach to comprehending how these structures respond to seismic events, thus enhancing safety assessments and enabling informed decision-making. However, SHM for masonry structures remains exceptionally challenging due to the aforementioned complexities. Traditional SHM methods, reliant on global response data, often struggle to pinpoint damage effectively. Concurrently, externally attached off-the-shelf sensors prove inadequate for monitoring masonry structures with extensive and intricate geometries. Such sensors typically provide localized information that does not adequately represent the broader structural response evolving within structural macro-elements. To surmount these challenges, novel sensing strategies are imperative to establish a more direct connection between damage assessment and decision-making. In this context, author’s research group at the University of Perugia, Italy, is pioneering innovative strain-based SHM techniques for existing masonry structures. The key approach involves the deployment of "smart bricks," which are strain-sensing micro-composite clay bricks capable of generating electrical outputs when subjected to external loads. These smart bricks replace conventional bricks at specific locations, supplying valuable data for damage identification through average masonry strain comparisons under permanent loads before and after seismic events. This work highlights group’s recent achievements in sensor development, encompassing sensor fabrication, characterization, and damage identification. Additionally, the talk will delve into the journey towards achieving a truly intelligent masonry by integrating smart bricks with smart mortar layers, providing insights into the detection of cracks and larger-scale structural damage within these heritage structures.

The next generation of space telescopes will require large, segmented apertures for observations in the near ultraviolet through mid- and far-infrared regions to enable new science ranging from exoplanet characterization to precision astronomical observations that refine astrophysics models. To meet these challenges, we are developing instrumented (strain gauge) surface parallel actuators (SPAs) that are robust and can meet the stringent requirements of mass and cost per m². We have developed a surface parallel mirror test piece and a set of flexured actuators that maintain compression in the piezoelectric stack elements at all times. The characterization work of these actuators is directed at understanding the performance of flexure piezoelectric multilayer stack actuator operation when embedded in the mirror. To determine the influence functions for each actuator position, we will report the measured stroke/strain and charge/capacitance versus voltage curves for all 42 preloaded actuators. Although designed to operate under close loop control via feedback from the strain gauge initial testing on bare lead zirconate titanate (PZT) stack actuators suggests that by driving the stack to a known domain state we could perform open loop control in the actuators to levels of ± 0.3 μm. We will also report on creep for the actuators and cross actuation for each unique actuator position as well as discuss approaches to mitigating the effect on open loop control error. Thermal studies of flextensional actuators embedded in analog rib structures down to 100K will also be presented.

The conventional method for one-dimensional acoustic emission (AE) source localization relies on two sensors to calculate the time-of-flight (TOF) and subsequently determine the source location given the wave speed. However, this method encounters inaccuracies when applied to dispersive mediums. To address this challenge, we suggest a novel method for single-sensor source localization. This method leverages multi-frequency Micro-Electro-Mechanical Systems (MEMS) equipped with sixteen resonators, each tuned to a specific frequency within the 100 kHz to 700 kHz range. By employing the wavelet transformation technique, we can determine the arrival times of these sixteen unique frequencies. By incorporating these arrival times with the dispersion curve, obtained from numerical simulation, it becomes possible to pinpoint the acoustic emission source location using just one AE sensor. Initial experimental validation conducted on a steel plate demonstrated the method's validity, achieving approximately 90% accuracy in source localization. This technique not only streamlines the process by eliminating the need for multiple sensors but also provides reliable source localization results in dispersive mediums.

Radiation-induced attenuation can pose a great challenge to the implementation of optical sensors in extreme environments. Photobleaching is known to mitigate the damage caused by radiation but a lot is yet to be investigated. In this work, we look at the power-dependence of the photobleaching phenomenon at cryogenic temperature. We used three standard fibers carrying around 2mW of light at 1550nm and 4, 0.4, 0.09 of the light at 1050nm respectively; and a fourth standard control fiber that carried no photobleaching light. We observed a large reduction in radiation-induced attenuation in all of the fibers with light at 1050nm when compared to the control fiber. This reduction, however is not linear and saturates for higher powers. These results are consistent with our theoretical models.

Tendon damage is a major risk of prestressed structures. Ground anchors are structural elements that introduce high levels of prestress, typically over 1000 kN. Tendon damage can threaten the stability of the structure they support. Particularly, visual inspection of buried ground anchor tendons is impossible. Thus, assessing tendon damage is essential during its service life. This study proposes an embedded tendon damage detection method based on the magnetic hysteresis change. Different damage degrees in tendons were measured at room temperature using the fabricated electromagnetic induction (EMI) sensor. Subsequently, the induced electromotive force (EMF) and magnetic flux density were obtained. The finite element simulation results showed a quadratic relationship between the effective cross-sectional area reduction ratio of the specimen due to damage and the peak of magnetic flux density. The experiment results were compared with the simulation results. This study introduces a promising nondestructive evaluation (NDE) method for detecting damage in the embedded tendon of ground anchors.

Safe storage of spent nuclear fuel (SNF) is critical to the nuclear fuel cycle and the future of nuclear energy. In the United States, SNF is primarily stored via two methods regulated by the U.S. Nuclear Regulatory Commission (U.S. NRC): wet storage in SNF pools, and dry storage in dry cask storage systems (DCSSs). After about five years of cooling in spent fuel pools, the fuel assemblies are transferred into DCSSs, and the systems are filled with helium and sealed by welding. Deterioration of conditions inside of a DCSS will be reflected by changes in the internal gas properties which motivates the development of acoustic techniques to monitor internal gas properties, over extended storage periods, using sensors mounted on the exterior of the storage packages. However, a major challenge in collecting acoustic signals is the impedance mismatch between the steel canister shell and the gas. Only a small fraction of the ultrasonic signal can be transmitted through the gas medium. In this paper, experimental studies on a full-scale canister mock-up were conducted to capture the gas-borne signals. Damping materials were pasted on the outside and blocking and unblocking tests were conducted to identify the gas-borne signal. The results showed that the excitation frequency plays an important role in maximizing the gas-borne signal. The gas-borne signal was successfully detected at around the theoretical time-of-flight (TOF). A high signal-to-noise ratio (SNR) was achieved in the measurements. Next, the acoustic impedance matching (AIM) layers were introduced, and the gas signal was drastically improved compared with no AIM layers.

In this study, we harnessed the potential of ensemble machine learning models and artificial intelligence (AI) techniques to address the limitations of conventional scour depth estimation methods. Our primary objective is to investigate how these advanced techniques can improve the accuracy of scour depth predictions, enabling more effective bridge engineering and maintenance practices. To achieve this objective, we employed ensemble machine learning (ML) approach comprising CatBoost, random forest, histogram gradient boosting, and extreme gradient boosting algorithms. These models were rigorously trained and tested on field data to predict bridge scour depth accurately. Furthermore, the performance of the ensemble machine learning models was benchmarked against that of an Artificial Neural Network (ANN) model. The key findings of this study show the superior performance of ensemble ML models over the ANN model in predicting bridge scour depth. Among the ensemble models, the CatBoost algorithm had the best performance with R-square value of 0.85 and a minimal root mean square error (RMSE) value of 0.47.

Palladium thin films have been studied as hydrogen sensing materials and applied to variety of optical hydrogen sensors. Recently, tantalum has emerged as an attractive option for hydrogen sensing materials due to its broad sensing range and flexibility in tuning the sensing range by modifying the alloying composition or elements. Following the demand for optical hydrogen sensors for aerospace applications, testing the performance of hydrogen sensing materials is of interest. This work examines the optical response in respect to changing hydrogen concentrations and thermal expansion of palladium-gold (Pd_0.65Au_0.35) and tantalum-ruthenium (Ta_0.97Ru_0.03 and Ta_0.91Ru_0.09) thin films at temperatures similar to a hydrogen combustion engine. Our results suggest that tantalum-ruthenium alloys are suitable for sensing hydrogen from ambient temperatures up to 270°C because its low detection limit (0.01% of hydrogen in the atmosphere) is well below the explosive limit of hydrogen (4% of hydrogen in the atmosphere).

This study presents a multi-modal artificial intelligence (AI)-based water pipeline maintenance method based on RGB images and ultrasound data. Our methodology leverages the concurrent collection of visual and auditory data from pipelines to improve the detection and prediction of anomalies such as abnormal welds, corrosion, scale, and cracks. By converting ultrasound data into spectrogram images using short-time Fourier transform (STFT) and combining them with RGB images, we create a composite data input for a convolutional neural network (CNN) model. This model is trained to classify the condition of water pipelines into distinct categories based on multi-modal inputs. The fusion of these two data modalities aims to significantly enhance the accuracy of pipeline anomaly detection, offering a novel approach for predictive maintenance in water pipeline facilities. We tested on 6 different classes with each 100 pair of datasets, therefore a total of 600 pairs of RGB and spectrogram images, and achieved an average accuracy of 92.7%. Our research contributes the potential application of multi-modal AI in pipeline maintenance.

Optical wavefront analyzers (WFS) play an important role in optical metrology. Previously, our team developed an optical wavefront analysis system called SPARROW (shear phase analyzer, reflective/reflective optical wavefront), which utilized a right angle phase-shifting mechanism and a thermal resistance wire actuator to integrate opto-mechanical systems for wavefront measurements. The feature of the innovations was its compact size. During the measurement process, the whole SPARROW can be rotated by 90 degrees to obtain the wavefront gradients in two orthogonal directions which prevent it from being used for real-time measurements. To overcome this disadvantage, a beam rotation optical mechanism was proposed to replace the required step of the SPARROW rotation. The un-tested wavefront beam was polarized first and then split into two similar polarized beams. One of the beams along with its polarization was rotated by 90 degrees and the two orthogonal polarizations beams each carried information about the two wavefront directions. Then the rotated and unrotated beams were combined into one beam and it became incident into the phase-shifting mechanism to complete shear interferometry. Finally, the use of a PBS (polarizing beam splitter) separated the beams to achieve wavefront gradient measurements in two orthogonal directions. Therefore, we can reconstruct the wavefront with gradient data and undertake analysis. In addition, the improvement to the reconstruction process by the improvement of the thermal resistance wire actuator can reduce the time required for a single measurement cycle. With a combination of these efforts, a fast wavefront measurement can be obtained.

In this study, the creation of a 3D digital model of a bridge structure by applying computer vision technology to LiDAR point cloud data acquired through laser scanning is proposed for determining structural damage. This method first identifies and labels the individual components of a bridge and then creates a detailed 3D digital model of its structure. An octree data structure is used to efficiently store and process the point cloud data to create the digital model. A static loading test on an actual bridge structure demonstrated the effectiveness of the proposed method in identifying individual components and analyzing the structural deformation. Based on the results, a 3D digital model can be used to identify structural damage to bridges during bridge inspection, monitoring, and maintenance.

Expanding the capabilities that are used in the NASA’s exploration of planetary bodies in our solar system would require mechanisms and actuators that can operate at cryogenic temperatures (-240 °C to -60 °C) in dusty environments. These applications include the exploration of lunar surface environments with temperatures that are below -100 °C. For this purpose, the authors are working on developing piezoelectric motors capable of operating at such extreme conditions. Novel piezoelectric motors were conceived and are being investigated to enable precision deployment and motion mechanisms that can be used for potential actuation of antennas and solar arrays, lower power robot arms, and percussive drills. This motor technology is intended to be integrated in a testbed developed at NASA to demonstrate its capabilities once it has been characterized at room temperature. These motors are being developed as game changers for enabling rotational drive mechanisms (rovers, robots, gimbals, drills, etc.) in extremely cold and dusty environments. These drive systems will be operated without the use of heaters or atmospheric control chambers (which eliminates grease lubrication) to raise the actuator’s temperature. Further, these motors will enable actuation of very high precision mechanisms having lower power motion without gears or gear lubrication, backlash, or power consumption to hold position. These actuators contain piezoelectrically-excited fixtures that are vibrated out of phase such that they sequentially push the rotor to produce continuous rotation. A proof-of-concept linear actuator that uses fixtures with flexure-preloaded piezoelectric stacks and operates in an inch-worm configuration at low frequency has been developed and demonstrated. Further, a proof-of-concept rotary actuator is currently being developed that uses a V-shaped piezoelectric fixture driven in resonance that generates an elliptical motion at the horn tip to drive a rotor. In this paper, the latest progress will be presented.

This work explores the sensing potential of mycelium with the intention of incorporating this as an intrinsic sensing mechanism within structural materials. Infrastructure plays a critical role in modern societies with regard to economic productivity, social cohesion, and community well-being. By merging materials that are used for construction, such as concrete with living components, we aim to add intrinsic monitoring mechanisms that could usher in a new era of structural monitoring solutions. Mycelium, the vegetative part of the fungi, has been shown to have an extracellular electrical potential that changes when exposed to various physical and chemical stimuli, making it an ideal candidate for this purpose. In this preliminary investigation, we analyse the electrical behaviour of mycelium exploring its potential use as a sensing material within infrastructure components.

Structural Health Monitoring (SHM) plays a vital role in maintaining the integrity of structures by providing continuous information about damage or anomalies. Vibration-based SHM, which focuses on the dynamic behavior of structures, offers insights into structural conditions through changes in dynamic properties. Among SHM approaches, damage localization is crucial for pinpointing the geometric location of damage. This paper proposes a method for damage localization using Short Time Fourier Transform and a Statistical Interpolation Damage Index. The proposed methodology is applied to a numerical case study involving a finite element beam model and to the S101 benchmark bridge, in Austria, demonstrating its efficacy in damage localization. The study also introduces a multi-level clustering approach to perform damage localization using smart decentralized sensor networks, able to reduce the volume of transmitted data and thereby the energy requirements. Results show promising outcomes in accurately identifying damage locations while minimizing data transmission.

In this paper, we report a method to control the rotational action of a planer-type piezoelectric motor. The motor is a planer-type piezoelectric actuator that uses two bending modes to create multi-directional traveling waves. The structure is simple and cost-effective. The multi-integer Frequency, Two-Mode method is used to drive the x- or y-directors, and the One-Frequency Two-Mode method is used to drive the motor rotates with respect to the z-axis. To control the multi-directional motorization of this motor in the x and y directions, we integrate a gyroscope to perform a precision directional control with feedback control. The gyroscope senses the rotational signal with respect to the z-axis, and the angular rotational signal is used to monitor the movement of the motor. The design and experimental studies on the multi-directional piezoelectric linear motor are discussed in this paper.

Multiple scattering is a common phenomenon in acoustic media that arises from the interaction of the acoustic field with a network of scatterers. This mechanism is dominant in problems such as the design and simulation of acoustic metamaterial structures often used to achieve acoustic control for sound isolation, and remote sensing. In this study, we present a physics-informed neural network (PINN) capable of simulating the propagation of acoustic waves in an infinite domain in the presence of multiple rigid scatterers. This approach integrates a deep neural network architecture with the mathematical description of the physical problem in order to obtain predictions of the acoustic field that are consistent with both governing equations and boundary conditions. The predictions from the PINN are compared with those from a commercial finite element software model in order to assess the performance of the method.

This paper introduces a Bayesian data fusion methodology for the monitoring of bridge displacements, employing a synergistic combination of satellite Interferometric Synthetic Aperture Radar (InSAR) and topographic measurements taken in free configuration. Focused on the case study of the Belprato 2 Viaduct, which is affected by a slow-moving landslide, this research demonstrates the potential of integrating diverse data sources to overcome the limitations posed by these monitoring techniques considered as alone. Our approach leverages the frequency and the remote, non-intrusive nature of InSAR technology and the accuracy of topographic surveys to obtain a high-resolution, three-dimensional bridge displacements caused by the landslide and temperature variations. The Bayesian framework facilitates the optimal fusion of these datasets, accounting for their respective uncertainties and different temporal resolutions. Moreover, it allows to include the information a priori on the landslide movements resulting for previous geological and geotechnical studies. The results from this study reveal significant improvements in the accuracy and reliability of displacement measurements, highlighting the benefits of data fusion for structural health monitoring. This paper highlights the importance of innovative monitoring solutions in the context of aging infrastructure, increasing environmental and traffic challenges, and complex topographical settings. Future directions for research include the exploration of real-time monitoring datasets and the integration of additional data types.

One of NASA’s priorities is the in-situ exploration of ocean worlds in the solar system where there may potentially exist life under the ice shell. This requires reaching the ocean below through great depths of ice at extremely low temperatures. Jupiter’s moon Europa is such a challenging body, whose ice shell is estimated to be 10’s of kilometers thick. An approach for reaching the ocean has been conceived using a melting probe “Cryobot” concept. A lander is assumed to be the platform from which the Cryobot would be deployed. This ice penetrating vehicle concept consists of a cylindrical, narrow-body probe that encases a radioisotope heat/power source that would be used to melt through the icy crust. The baseline design of the probe includes a suite of science instruments to analyze the ice during descent and the liquid ocean underneath. For wireless communication, which is the focus of this paper, acoustics and RF transceivers were developed as complementary systems. The RF is developed for use in the very cold porous top layer, while the acoustics is for communication in the warmer denser ice where dielectric absorption may preclude RF transmission. Acoustics/RF communication systems were developed, tested, and successfully demonstrated when frozen into a glacier at four points to transmit signals over a glacier ice distance of 120 m. The tests were conducted at the Matanuska Glacier, Alaska, about 70 miles northeast of Anchorage. The details of this study will be described and discussed in this paper.

The recent collapse of the I-95 overpass bridge in Philadelphia, PA, resulting from a fire incident, has underscored the need for a rapid evaluation of infrastructure for decision making process. To gain insights into the nature and magnitude of the damage, a section of deformed I-girders from the disaster site was acquired for in-depth examination. This research introduces a methodology employing a laser scanner to measure warping in the procured 4-foot segment of the girder. The girder was originally situated in the mid-span of the closest southbound girder to the fire that caused the northbound collapse. This novel scanning approach offers a thorough evaluation of deformations and warping patterns that can potentially be used for rapid evaluation of bridges after fire. In this way, this study aims to equip bridge inspectors and assessors with advanced tools, paving the way for more informed decision-making in future bridge-related accidents, including fire incidents.

The environmental thermal effects have a great influence on the structural behavior, while the Fiber Bragg grating (FBG) sensors used in the structural health monitoring (SHM) systems require thermal compensation. This paper presents a preliminary study on the thermal behavior of the prestressed double-T slab at the top floor of the Stadium Drive Garage at Princeton University. Embedded long-gauge strain and temperature sensors, based on FBGs, were employed to monitor the structural responses of the slab over a two-year period. The study aims to assess the temperature-strain relationship of the double-T slab with complex geometries and evaluate the long-term performance of sustainable carbon-capturing concrete, the material used in the construction of the garage. To enable the application of thermal compensation techniques, the calculations of the coefficient of thermal relation between the strain and temperature using measurement data from different time periods are carried out with the implementation of the mean thermal gradient (MeLG) method. Analysis reveals relatively stable values of the coefficient of thermal relation over the monitoring period. The findings of this study have important implications for understanding structural thermal behavior with complex geometries and new sustainable materials, and contribute to ongoing efforts to improve the resilience and sustainability of civil infrastructure.

This paper presents a magnetorheological damper applied to suppress vibration of a three-story building. The damping force is obtained through the Buc-Wen model, and then an inverse damper model is built up by the adaptive neuro-fuzzy inference system (ANFIS). To facilitate the vibration suppression, a semi-active control is designed, where a PID control is used to evaluate the desired damping force based on the building vibration and then the ANFIS model is used to calculate the desired damper current. The current is applied to the damper so as to generate a damping force to suppress the vibrations. The proposed approach is demonstrated through numerical simulations.

In the realm of multi-spacecraft missions, crew transport and satellite tasks require precision in rendezvous maneuvers. A robust navigation system becomes essential for addressing uncertainties in space robotic modeling. This study presents a novel approach by leveraging neuromorphic computing, introducing the Spiking Neural Network-Modified Sliding Innovation Filter (SNN-MSIF) for satellite rendezvous in circular orbit. The SNN-MSIF combines the efficiency of neuromorphic computing with MSIF's robustness, enhancing accuracy and stability. Utilizing Clohessy-Wiltshire equations, the model captures relative motion between spacecraft. Monte Carlo simulations are used to compare the SNN-MSIF with SNN-Kalman filters and their non-spiking counterparts, showcasing the superior accuracy and stability of our approach. The evaluation of their robustness under uncertaintie1s and neuron silencing demonstrates their reliability. The findings establish SNN-MSIF as an effective, efficient, and promising filtering framework for space robotics, refining navigation, and addressing multi-spacecraft challenges.

In the realm of structure health monitoring for pressure vessels intended for space habitats, identifying sensor anomalies is of critical importance. The sensor anomalies are data patterns that diverge from anticipated measurement behaviors. To address the multifaceted challenges, we propose a hierarchical mechanism for sensor anomaly detection. This strategic approach not only filters out aberrant data but also subsequently ensures the extraction of reliable results for structure health monitoring, providing a safeguard against potential erroneous decision-making. Furthermore, this approach allows for efficient data handling across multiple sensors and incorporates physical knowledge into the deep learning model to comprehensively detect any sensor anomalies that are physically implausible. As a result, we achieve a more holistic and robust detection of sensor anomalies, ensuring heightened reliability in health monitoring for pressure vessel.

One of the main issues facing reinforced concrete structures, which are the largest category of infrastructure, is corrosion. Corrosion of metallic reinforcing in concrete is influenced by a variety of attributes of the concrete, including depth of cover and the porosity of the hardened or mature concrete. Ground penetrating radar (GPR) is a leading tool for determining and verifying cover depth and other physical features like delamination. It is an emerging tool for characterization of the concrete material itself, including applications to study hydration processes and early age properties. Recent work has identified relationships between GPR signals and mature concrete porosity measured using cold-water saturation. This work presents an initial study that compares GPR attributes and physical properties (cold water saturation porosity, compressive strength, and density) to new measurements of the porosity performed using the vacuum saturation method. This method is applied to a small set of mature samples and the results are compared across each measurement and sample characteristic. On average, all mixes except concrete paste (containing no coarse aggregate) have higher vacuum saturation porosity than cold water saturation porosity because the method captures smaller pore sizes. Basic GPR attributes (maximum amplitude, total energy, and average amplitude) have no strong direct linear correlation with either porosity measurement, but there are some promising relationships found in the average trends. This study gives insight into the porosity-controlled mechanisms of GPR signal propagation and how they may be used to evaluate mature in situ concrete without collecting or directly testing samples.

The Smart Gear System comprises two integral components: the Smart Gear, featuring a sensor and an antenna circuit intricately embedded onto the gear, and a monitoring antenna seamlessly linked to a network analyzer. The proximity of the smart gear antenna to the monitoring antenna induces magnetic coupling between the two antennae, consequently altering the frequency characteristics of the monitoring antenna. Additionally, minute deformations in the gear tooth root result in modifications to the sensor's shape on the gear, thereby causing alterations in the frequency characteristics of the sensor. The overall health status of the gear can be discerned on the monitoring antenna side by monitoring the frequency characteristic variations of the sensor, based on the fundamental principle of magnetic coupling during operational processes. In the context of this investigation, a dynamic evaluation of the Smart Gear System is executed, with the return loss of the monitoring antenna meticulously gauged at periodic intervals throughout the operational phase. This study delves into the correlation between the return loss exhibited by the monitoring antenna and the dynamic physical alterations in the Smart Gear during operation.

In view that cities will continue to house the majority of the world’s population at an increasing rate in association with the face of climate change, we quantify urban resilience by examining the response history of the mean-square displacement of the citizens of large cities prior and upon historic natural hazards strike. The recorded meansquare displacements of large numbers of cell-phone users from the cities of Houston, Miami and Jacksonville when struck by hurricanes Harvey 2017, Irma 2017 and Dorian 2019, together with the recorded mean-square displacements of the citizens of Dallas, and Houston when experienced the 2021 North American winter storm, revert immediately to their pre-event steady-state response; suggesting that large cities when struck by natural hazards exhibit an inherent resilience. We explain how the mean-square displacement from a random (stochastic) process is intimately related to deterministic time-response functions of emergent mechanical models. We build on this overarching relation that derives from Langevin dynamics and we show that a significant number of records presented in this study validate a mechanical model for cities, recently developed by the authors. Our mechanical model that is inspired from the quantitative theory of Brownian motion predicts that following a natural hazard, large cities revert immediately to their initial steady-state regime and resume their normal, pre-event activities-that is exactly what the recorded data show.

Pipeline pigging has become a standard practice in the petroleum and natural gas industry, utilizing fluid or gas pumping upstream to facilitate cleaning of wax, sediments, or dewatering. In previous research by the authors, a specialized epoxy membrane embedded with chemical sensors was developed to monitor in-line environments. However, the durability of this thin film material under pipeline cleaning conditions remained unexplored. This study conducted experimental tests simulating industrial pipeline cleaning procedures using rubber-based cleaning disks. Results indicate that a standard 3mm epoxy base attached to the pipeline's inner wall maintains integrity with minimal surface wear during pipeline cleaning activities. Additionally, the study found that the maximum shear stress applied to the membrane increases with thickness, reaching a maximum of 2.95 MPa when the membrane thickness is less than 3mm. These findings provide valuable insights into the resilience of epoxy-based membranes in harsh pipeline cleaning environments, informing future development and deployment of in-line monitoring technologies.

In this study, we aim to characterize the mechano-luminescent (ML) light emission of the copper-doped zinc sulfide (ZnS:Cu)-embedded polydimethylsiloxane (PDMS) (ZnS:Cu-PDMS) micro-composites. First, the seven different ZnS:Cu-PDMS micro-composites were fabricated by using three different types of ZnS:Cu that emits blue, green, and orange light by blending those different colored ZnS:Cu. The micro-composites were prepared through the mold-casting method. Second, the ML test specimens will be subjected to uniaxial tensile loading/unloading cycles with various loading patterns. The ML light emissions were recorded by a high-speed color camera, and the video footages were image-processed to catalogue ML light intensity and color. Lastly, the spectroradiometer was employed for acquiring the light radiance spectrums from the prepared ZnS:Cu-PDMS micro-composites.

Ingress of chlorides into reinforced concrete elements results in corrosion of reinforcing steel and loss of service life; new evidence also suggests that chlorides can also result in the deterioration of concrete itself. Estimating the rate of chloride ingress in concrete is difficult because current methods depend on measuring concentration profiles using chemical methods. In this paper, we proposed that EIT (Electrical Impedance Tomography) can be used to monitor the diffusion of chlorides into concrete since the presence of chlorides increases the electrical conductivity of the concrete.

The concept of using a capacitive-based sensing skin for structural health monitoring applications was proposed more than a decade ago. The sensing principle is to form a dense sensor network by assembling numerous soft elastomeric capacitors (SECs) in a matrix form, with each SEC acting as an independent strain gauge, resulting in a sensor network measuring local strain information over a global area analogous to biological skin. The SEC technology has considerably evolved since it was first conceptualized in 2009 in order to facilitate field implementation. For instance, the polymer mix and sensor design were modified to increase robustness and sensing performance, dedicated electronics was fabricated to improve resolution and accuracy, and various signal processing algorithms were formulated to transform measurements into actionable information. The objective of this paper is to discuss the evolution of the SEC technology starting from its conceptualization to its field validation in order to support and stimulate other research in sensor development geared towards structural health monitoring (SHM) applications. This includes discussions on 1) materials development; 2) signal acquisition and characterization; and 3) bridging the gap to field deployment.

A present challenge in structural health monitoring consists in the detection, localization, and quantification of small damage (e.g., small cracks) within large structures, such as bridges and buildings. Existing sensing solutions have several limitations, the most important being those related to the extent of spatial coverage by sensors and power supply. In this work, we will present proof-of-concept research for sub-millimeter displacement measurement using novel embeddable passive wireless radio frequency (RF) sensors. The novel sensors estimate relative displacement from phase shifts in the transmitted RF signal. The proposed system represents a novel paradigm in wireless sensing in structural health monitoring, as the wireless sensors are battery-less and will be deployed in a form of densely populated 3D network embedded within large volume of material.

The passively reconstructed Impulse Response Function (IRF) (or Green’s function) between two points of a medium can return a wealth of information about the dynamic behavior of the system including the existence of defects. For a linear system subjected to an unknown excitation source, the deconvolution operation between two monitored points can properly reconstruct the IRF between the two points, effectively using one of the points as a “virtual” source. This technique has been successfully used for studying the global dynamic behavior of structural systems, such as buildings and bridges, in the low frequency range. In civil engineering this approach is often referred to as Seismic Interferometry (SI). Various studies using SI have discussed the high frequency regime of transient wave propagation. This paper will focus on this high frequency regime and propose some novel implementations of SI to detect local structural damage in multistory buildings. Results will be presented from an aluminum rod and a laboratory-scale multistory aluminum frame.

In this paper, we report a piezoelectric rotational motor that can perform precision angular motorization control. This rotational motor is a composite composed of a stainless-steel plate and four PZT actuators, and mathematical model and numeric analysis were employed to determine optimal parameters for clockwise and counterclockwise rotations. The driving method is based on activating opposing bending modes with identical driving frequency but different phase. Hall sensors and magnet ring are integrated to provide a PID control for precision control. An accuracy of 0.145 deg is realized.

The use of strain gauges is foundational to structural health monitoring, allowing infrastructure to continuously observe strain, infer stress, and potentially detect fatigue/fracture cracks. However, traditional strain gauges have drawbacks. In addition to being costly, a single-element strain gauge will only detect strain in a single direction and must be mounted on smooth surfaces to ensure good adhesion. Soft Elastomeric Capacitors (SECs) have been proposed as a low-cost alternative to traditional strain gauges while allowing for a broader range of applications. They are flexible and can be modeled with different dimensions based on the monitored structure. Each SEC consists of three layers; the two outer layers act as electrodes and are made of a styrene-ethylene-butylene-styrene polymer in a matrix with carbon black. The inner (dielectric) layer comprises titanium oxide in a matrix with SEBS. The use of the SECs is not limited by the geometry of the surface being monitored, and it can, therefore, be adhered to a variety of surfaces as its flexibility allows it to conform to the irregularity and complexity of the monitored structure. The change experienced by a structure will correlate directly to the change in capacitance observed across the sensor, which can be used to predict the monitored structure’s state. While SECs have been studied for applications on various materials, experiments have been limited to adhering the sensor to smooth surfaces. However, concrete structures have various surface finishes that are not uniform, often deriving from an architect’s aesthetic desire. This work tests a corrugated SEC through compression tests on concrete samples with different surface finishing to investigate the effect of surface finishing on the SEC-measured strain. Each concrete sample is subjected to loading by a dynamic testing system, and the data collected from the SEC are compared to off-the-shelf resistive strain gauges. The results show that the performance of the cSEC on the different surfaces is not hindered by different concrete finishes, where a high signal-to-noise ratio of 21 dB and low mean absolute error of 22 μϵ is seen on the concrete specimen with a rough concrete surface. The strain metrics and surface effect on SEC performance are discussed.

High-temperature operation of vessels and piping systems in advanced nuclear reactors makes them susceptible to creep damage. To address this issue, the acoustic emission (AE) method was investigated as a structural health monitoring solution for detecting the initiation and evolution of creep damage. Stainless steel coupons were tested at 650°C to induce different creep mechanisms. Two AE sensors functional at room temperatures with a frequency bandwidth of 100-300 kHz were attached with waveguides, which were welded to the coupons to transmit and minimize the signal loss between the coupon and the waveguide. Once the AE characteristics of creep damage were identified, they were compared with the attenuation curve of an actual piping network at Argonne National Laboratory, Mechanisms Engineering Test Loop (METL) facility. The liquid sodium piping system at the METL facility was instrumented with twelve piezoelectric sensors strategically placed using waveguides to accommodate the high operational temperature. The AE simulations using pencil lead break (PLB) testing showed that the piping system operates as an excellent waveguide. The AE waveforms obtained from the laboratory setup and the attenuation curve of the actual piping system were analyzed to translate the laboratory-dependent AE data for creep damage to a realistic condition.

Ultrasound attenuation plays a vital role in mechanical engineering for material characterization and non-destructive testing. Determining the attenuation parameter can be challenging in applications involving microstructural changes. This paper introduces a novel method for determining the ultrasound attenuation parameter using fringe spectrum analysis from a UFPR, enhancing the sensitivity in detecting attenuation parameter shifts with microstructural changes in aluminum alloy and providing a quantification of these shifts by employing a metric derived from standard method. The UFPR, drawing on principles from optical/microwave Fabry-Perot interferometry, offers fringe frequency domain analysis as an alternative to conventional time-or-frequency-domain analysis, providing deeper insight into material properties. Our research studied the sensitized aluminum alloys that the Al-Mg particles precipitate at the grain boundaries due to prolonged heat exposure, which is ambiguous with the ultrasound attenuation parameters determined from conventional methods.

This paper aims to investigate the implementation of bonded UFPR using an optical fiber as the ultrasound waveguide. A longitudinal UFPR is created by bonding a section of an optical fiber with an FBG on an aluminum plate. The experimental setup consists of a broadband ultrasound transducer to excite broadband longitudinal waves into the structure and an FBG-based ultrasound sensing system to measure the resonance of the bonded UFPR. A mechanism to bond the optical fiber to the structure with controlled length and thickness has been developed. Time-domain and frequency-domain analyses confirm UFPR resonances and reveal interference peaks, supporting the principles of the FPR. This work also explores the effect of adhesive length on resonance characteristics.

This study introduces an advanced Adaptive Optics (AO) technique employing a Digital Micromirror Device (DMD) for effective wavefront correction. The innovation lies in utilizing the DMD as a Spatial Light Modulator (SLM), marking a departure from traditional methods that mainly used Liquid Crystal on Silicon (LCoS) for this purpose. With its high resolution (912×1440 pixels) and ultra-fast response, the DMD enhances the system's efficiency in correcting optical aberrations. By applying the Lee Hologram Method within off-axis binary holography, our approach generates inverse wavefronts to counteract environmental distortions, thereby improving image clarity. The integration of a DMD-assisted Lateral Shearing Interferometer, which adjusts the incident light angle through wavefront modulation, accelerates optical path difference (OPD) measurement. Implementing a Five-Step Phase Shifted Method for both X and Y directional fringe patterns facilitates rapid phase retrieval. This process, crucial for reconstructing and correcting wavefront aberrations, leverages the differential phase information from the Lateral Shearing Interferometer. Our experimental results affirm the system's effectiveness in not only rectifying wavefront distortions but also in showcasing the enhanced performance capabilities brought forth by the DMD-assisted Lateral Shearing Interferometer.

Person identification is important in smart buildings to enable personalized services, such as monitoring individuals’ gait health. Existing studies found that the structural vibrations induced by human footsteps provide both identity and gait health information of individuals, such as a person’s walking speed, balance, and symmetry, enabling personalized gait health monitoring in smart buildings. However, footstep-induced structural vibrations not only depend on human walking patterns but also on a person’s footwear as the footstep force transmits from the foot to the floor. This co-dependency leads to difficulty in identifying the owner of the footsteps when multiple people share the same space and each person has multiple pairs of footwear. In this study, we characterize the effect of footwear on footstep-induced structural vibrations to recognize individuals even when they wear different pairs of shoes (or barefoot). We develop a new metric named Force Transmissibility (FT) that measures the proportion of forces transmitting from the foot to the floor through the footwear. This metric unifies the effect of diverse shoe types, and we utilize this metric to enable robust person identification among various shoe types. We evaluated our approach through real-world walking experiments with eight shoe types shared by four participants. Our method achieves a 22% improvement in identifying the owner of the footsteps when compared to a baseline without footwear considerations.

We investigate flexure piezoelectric stacks to produce large force and fine position control. The network model for the free ideal stack with many layers was first worked out by Martin [Martin 1963, 1964a, 1964b]. He noted that the stack impedance model in the limit of large layer number n > 8 that a network model for this case could be developed which is identical in form to the network model of the length-thickness mode with appropriate material coefficients from the length extensional (LE 33) mode of the material. This network model allows for the additional determination of the velocities and displacements of the stack surfaces, as well as the acoustic power deliver to any elastic load. We show that this model can be extended to allow for the modelling of acoustic elements, such as non-piezoelectric endcaps and coupling to other structures including flexures. In order to demonstrate the utility of this modeling, we will present the model for a stack embedded in a simple flexure frame. Embedding the piezoelectric stack in a flexure enables preloading of the piezoelectric so that it does not experience tension during operation. This reduces the overall risk of failure. Additionally, flexures have been demonstrated to amplify in the transverse direction the stroke of the stack at low frequencies by a factor that is proportional to the cotangent of the liftoff angle – the angle the flexure makes with the axis of the stack. Impedance measurements of the flexure stack show two additional modes in addition to the stack fundamental length extensional mode. Investigations on the other modes of the flexured stack actuator show a low frequency flexure mode that is π out of phase with the stack extension and with a resistance at resonance that is smaller by the amplification factor of the flexure. A flexure breathing mode is found just below the stack resonance. At the higher frequency a clamped stack resonance in phase with the flexure displacements is shown. In the paper, we will also discuss how the flexure network model can be implemented into other structures.

This paper introduces a methodology for optimizing 4D printing design through the integration of Residual Neural Network (ResNet) and Genetic Algorithms (GA). Departing from traditional forward design approaches, our inverse design methodology addresses both the forward prediction and inverse optimization problems. ResNet efficiently predicts the performance of 4D-printed parts given their design, while GA optimizes material allocation and stimuli distribution to achieve desired configurations. The ResNet model exhibits high accuracy, converging to a small error (10⁻³), as validated across diverse cases. The GA demonstrates effectiveness in achieving optimal or near-optimal solutions, illustrated through case studies shaping parts into a parabola and a sinusoid. Experimental results align with optimized and simulated outcomes, showcasing the practical applicability of our approach in 4D printing design optimization.

With the development of the Internet of Things (IoT) technology, a large number of wireless devices are deployed in our surroundings. The power supply issue of these IoT devices has become increasingly noteworthy. On the other hand, the installation and maintenance of a large number of widely distributed sensors are cumbersome. Taking the smart floor for example, in this article, we introduce a battery-free wireless floor tile design based on piezoelectric energy harvesting (PEH). When being excited by a footstep, the piezoelectric floor tile generates an impulsive voltage, which needs to be properly converted to a stable DC form. An energy management circuit is also necessary for properly carrying out the energy buffer-release procedures. Therefore, the circuit part includes components such as an AC to DC rectifier, an under-voltage lockout module, and a voltage regulator. The obtained stable DC voltage is then used to wake and power the Bluetooth node to properly function. Although there were many similar PEH floor tile designs previously built with the same low-cost piezoelectric buzzers, this design is the most compact and inclusive, as a self-powered Bluetooth beacon module. Compared to many energy harvesting floor tile designs, this one has another benefit of a small stroke distance, less than 1 mm, which brings little discomfort to natural walking. Given its motion-powering feature, the power-supply issue, easy deployment, and maintenance issues are all solved. Such smart pavement has the potential to further provide useful information, such as crowdedness, individual tracking, and fall detection, about pedestrians.

Computer vision and signal processing techniques have been adopted in several areas of human knowledge. Usually, these techniques are applied to the extraction of information from images, in order to reduce common errors inherent in repetitive work and reduce production times. In industry, signal processing is used to inspect the quality of products, checking their irregularities and whether they are within acceptable tolerances, in accordance with quality standards. In civil construction pattern tracing and image processing is used with little or no frequency to monitor civil engineering structures. Applications of intelligent sensory systems and intelligent structures that enable traceability by image are the great technological innovation implemented in this study in the field of civil engineering. Wall cracks, fissures and other types of changes are pathological manifestations of buildings observed in masonry, beams, pillars, slabs, floors and other elements, usually caused by tensions in the materials. If the materials are requested with an effort greater than their resistance, failure occurs causing an opening, and according to their thickness it will be classified as Cracks or fissures. Among the various ways to acquire images in buildings is the use of Drones that use different cameras for image acquisition. The use of a camera coupled to the Drone has been an alternative to track objects and places that are difficult to access, in the case of bridges, buildings and viaducts, where there is a need for observation in loco, the use of these devices facilitates the verification of restricted and difficult areas access. In this study, the Pixy camera was selected in order to verify its versatility in capturing artificial images that simulate cracks and fissures. The adopted experimental method consisted of image capture using the Pixy camera coupled to the Drone in a controlled laboratory environment. After capturing the image, an algorithm programmed in C++ language with data correlation capability was used to identify the type of cracks and fissures. The main result obtained from the research identified that there is a distance of 30 cm that corresponds to the approximation limit of the Drone with the Pixy camera of the civil engineering structures to be analyzed. The algorithm developed in C++ language to program the Pixy camera enabled remote sensing and traceability of patterns of cracks and fissures in civil engineering structures, with an accuracy of 99.99%, a result that corroborates the efficiency of the research method adopted.

In this research, we aim to address the critical issue of limited awareness among tribal communities regarding the advanced features of their connected vehicles in emergency situations. Our research involves conducting a comprehensive stated preference survey among tribal vehicle owners, assessing their awareness of connected vehicle functionalities and their comfort with technology. By gathering insights into their preferences and knowledge levels, we aim to inform transportation policies and strategies tailored to tribal regions, enhancing the safety and resilience of transportation systems in emergencies, ultimately benefiting tribal communities.

Civil engineering structures are routinely exposed to corrosive environments, posing threats to their structural integrity. Traditional corrosion control methods often involve employing physical barriers, such as various coatings, to isolate the steel substrate from surrounding electrolytes. Among these methods, thermal spraying of alloy coatings has emerged as a prominent technique in safeguarding steel matrices against corrosion, particularly in industrial and marine settings. However, the inherent porosity of thermal spraying coatings compromises their corrosion resistance. Incorporating a polymer top layer offers a promising solution by sealing pores and augmenting overall performance. This study investigates corrosion on duplex-coated steel utilizing distributed fiber optic sensors based on optical frequency domain reflectometry. Experimental analyses involve embedding serpentine-arranged distributed fiber optic strain sensors within both thermal spraying layers and epoxy layers. Results demonstrate the efficiency of distributed sensors in identifying corrosion propagation paths by measuring the induced strain changes. Furthermore, the duplex coating exhibits significant enhancements in corrosion resistance for steel structures.

Pipelines are an important type of transportation and are often buried underground, making regular maintenance and inspection challenging, especially for pipelines without network maps. Ultrasonic testing (UT) is a commonly used non-destructive method to assess surface or subsurface defects in pipeline. In this study, we proposed to correlate the UT signals with the geometric and spatial features to reconstruct the network of pipeline. The feasibility of the proposed method has been discussed numerically. Two key geometric features (pipe length and connections) were investigated to explore the correlation between ultrasonic guided wave features and different length and node conditions. This study used principal component analysis to select the characteristics, and integrated the backpropagation neural networks (BPNN) and radial basis function neural networks (RBFNN) to process the signals to establish the relationship between UT signal and spatial features. The results of the study show that BPNN performs better in pipeline length and connection type recognition, with an average coefficient of determination of 0.96 for recognizing the length and an average correct rate of 91.9% for recognizing the connection type. A comprehensive comparison of the two intelligent algorithms reveals that the BPNN performs well in improving the prediction of pipeline complexity, which significantly enhances the detection of geometric and spatial features of pipelines.

With the development of prefabricated building (PB), the connection quality problems among the prefabricated components may cause negative effect on the safety of the entire PB, and severe defects in the connection may lead to catastrophic accidents. Herein, the half-grouted sleeve connection is one of the most commonly used connection in PB While, the grouting process has to be completed on site; so, internal defects are unavoidable. Currently, there are few effective and easy-used inspection methods which can be used in field and realize the real-time monitoring. Therefore, this study proposes to use the acoustic emission (AE) method to monitor the damage progress of sleeves with different defect rates under monotonic tensile tension. Different defect levels were artificially introduced in the sample. Two different failure modes were identified by AE, and the AE signatures are different. The proposed method provides a feasible nondestructive method for prediction of damage in half-grouted sleeve in the early stage.

The folding-wing aircraft can obtain appropriate lift and drag by changing the folding angle of the wing. At different stages of the aircraft executing tasks, there can be a corresponding flight state. When the carrier -based aircraft is parked, the occupation space can be reduced by folding by the wing, which can increase the number of carrier -based aircraft. The folding wing drive mechanism is a key technology for folding the wing, which has a key impact on the characteristics of structural transmission. The driving force of the traditional folding wing mechanism is positively related to the size of the drive. Therefore, in the traditional folding wing drive mechanism, if the driving force required when the wing is folded is large, it must use a large size actuators, but this is often limited by the space at the wing shaft. To solve this problem, a folding wing auxiliary drive mechanism is designed in this article. This mechanism uses spring deformation to store the gravity of the wing unfolding process as elastic potential energy and release it when the wing is folded. The use of this mechanism not only increases the driving capacity of the folding wing drive mechanism, but also reduces the power requirement of the main drive. In order to convert the rotation movement when folding the wing into a linear motion, a rotary-to-linear device is designed in this article. In order to eliminate the restrictions on the working load and displacement itinerary in the spring design, this article designed a new type of energy storage spring device: a solid-liquid hybrid spring device, and two spring design schemes are given. Based on the demand of force and displacement strokes, this article gives a detailed design of the auxiliary drive mechanism, and a detailed description of the structural layout and specific operation method is given. On this basis, a key parameter of the auxiliary drive mechanism is given. Finally, the renderings of the auxiliary drive when the wing folding and unfolding are shown.

In this paper, a classification method of flutter test signals based on convolutional neural network (CNN) and Hilbert- Huang transform (HHT) is established, which can be effectively applied to flutter boundary prediction. This method combines convolutional neural network and time-frequency analysis. Firstly, the flutter test signal is preprocessed by Hilbert-Huang transform and labeled according to the actual signal source. The label contains channel information and flutter information. All the signals and labels are composed of the dataset, the dataset is randomly scrambled and 80% of the number is taken as the training set, and the features are extracted, and the classification model is trained by convolutional neural network. The remaining 20% of the dataset is taken as the test set. The test set is used to test the classification model and verify the reliability and accuracy of the model. The accuracy of the final test set is above 90%, which indicates that the model trained by this method can effectively identify the channel information and the flutter information of the signal.